Meta Eyes $10B Compute Deal to Power Anthropic AI
Meta is reportedly pursuing a potential $10 billion deal to lease computing power to Anthropic, a striking move that could reshape the AI infrastructure race and blur the lines between rival labs.
Moonshot's Kimi K3 Debuts as Largest Open-Weight Model
Alibaba-backed Moonshot AI has released Kimi K3, billed as the world's largest open-weight AI model, intensifying the open-source race and reshaping the competitive landscape that underpins generative and multimodal AI tools.
Deepfake Fraud Surges 180% as ID Checks Collapse
Deepfake-driven fraud attacks jumped 180% as synthetic media outpaces legacy identity verification systems, exposing critical gaps in KYC and biometric onboarding defenses.
Facial Motion Analysis Detects Deepfakes at 95% Accuracy
A new deepfake detection approach analyzes subtle facial movement patterns to identify synthetic videos with more than 95% accuracy, offering a behavioral-biometric alternative to pixel-level forensic methods.
OriginBlame: Tracing Data Provenance in AI Training
OriginBlame introduces a technique for tracing data provenance at both the record and token level in AI training datasets, offering new tools for accountability, attribution, and authenticity in the era of generative models.
Deepfake Abuse Surges as YouTube, X Fuel Discovery
A new report finds that mainstream platforms like YouTube and X are increasingly serving as discovery channels for abusive deepfake tools and content, accelerating harm and complicating detection efforts.
Reelful AI Turns Your Camera Roll Into Short Videos
Reelful launches an AI tool that automatically transforms photos and clips from your camera roll into polished short-form videos for social media, tapping into the booming market for automated content creation.
Polygraf AI Launches Real-Time Deepfake Meeting Guard
Polygraf AI's new Meeting Guard tool brings real-time deepfake detection to enterprise video calls, aiming to stop impersonation fraud and synthetic identity attacks before they reach decision-makers.
China AI Companion Law Wipes 8M Agents Off Top App
A new Chinese regulation targeting AI girlfriend and companion services triggered the removal of roughly 8 million AI agents from the country's largest AI app, signaling a major shift in how synthetic personas are governed.
Neural Audio Watermarking's Role in Audio Security
Neural audio watermarking is emerging as a key defense against voice cloning and synthetic audio. Here's how the technology fits into the broader audio security and content authenticity landscape.
Instagram Chief: Users Can Filter AI From Their Feeds
Instagram head Adam Mosseri says users who dislike AI-generated content should be able to keep it out of their feeds, signaling Meta's evolving approach to synthetic media curation and content controls.
OpenAI Debuts New Model Family Led by GPT-5.6
OpenAI has unveiled its latest generation of AI models headlined by GPT-5.6, marking another significant step forward in the company's foundation model roadmap and reshaping the competitive landscape for generative AI.